Apparatus, method, and recording medium

Abstract

An apparatus is provided, which includes an obtainment portion configured to obtain a plurality of images corresponding to respective values of a parameter that affects occurrence of a target event; a generation portion configured to generate, for each of a plurality of thresholds relating to the parameter, a learning data set in which at least a part of the plurality of images is each given an attribute value, the attribute value varying based on whether the threshold is exceeded; a learning processing portion configured to perform, for each generated learning data set, learning processing on an estimation model that estimates an attribute of an image in response to input of the image; and an evaluation value output portion configured to output, for each estimation model, an evaluation value according to an estimation error of the estimation model.

What is claimed is: 
     
       1. An apparatus, comprising:
 an obtainment portion configured to obtain a plurality of images corresponding to respective values of at least one of dosage, temperature, and exposure time of a drug that affects occurrence of a reaction of the drug; 
 a generation portion configured to generate, for each of a plurality of thresholds relating to the at least one of the dosage, temperature, and exposure time, a learning data set in which at least a part of the plurality of images is each given an attribute value, the attribute value varying corresponding to whether or not the at least one of the dosage, temperature, and exposure time exceeds the threshold; 
 a learning processing portion configured to perform, for each generated learning data set, learning processing on an estimation model that estimates the attribute value of an image in response to input of the image; 
 an evaluation value output portion configured to output, for each estimation model, an evaluation value according to an estimation error of the estimation model; and 
 a determination portion configured to determine a value of the at least one of the dosage, temperature, and exposure time at which the reaction of the drug occurs based on the threshold, among the plurality of thresholds, corresponding to the estimation model having the smallest evaluation value. 
 
     
     
       2. The apparatus according to  claim 1 , wherein
 the learning processing portion repeatedly performs learning processing on each estimation model using the corresponding learning data set; and 
 the apparatus further comprises 
 a determination portion configured to determine a value of the at least one of the dosage, temperature, and exposure time at which the reaction of the drug occurs based on a threshold, among the plurality of thresholds, corresponding to the estimation model in which the evaluation value becomes less than or equal to a reference value with the smallest number of times of learning. 
 
     
     
       3. The apparatus according to  claim 1 , wherein
 the generation portion generates the learning data set for each of the plurality of thresholds using images, among the plurality of images, corresponding to values of the at least one of the dosage, temperature, and exposure time, the values being within a range including the threshold. 
 
     
     
       4. The apparatus according to  claim 1 , wherein
 the obtainment portion obtains a plurality of images corresponding to respective values of each of the at least one of the dosage, temperature, and exposure time that affect occurrence of the reaction of the drug, and 
 the generation portion generates, for each of a plurality of combinations of thresholds relating to the at least one of the dosage, temperature, and exposure time, the learning data set in which each of the plurality of images is given an attribute value, the attribute value varying based on whether each of the thresholds included in each of the combinations is exceeded. 
 
     
     
       5. The apparatus according to  claim 1 , wherein
 the learning processing portion performs learning processing on each of a plurality of estimation models having the same initial condition, using a corresponding one of a plurality of learning data sets generated for the plurality of thresholds. 
 
     
     
       6. The apparatus according to  claim 1 , wherein
 the learning processing portion performs transfer learning using a model on which learning processing has been performed as the estimation model. 
 
     
     
       7. The apparatus according to  claim 1 , wherein
 the estimation model estimates the attribute value of an image in response to input of the image. 
 
     
     
       8. The apparatus according to  claim 1 , wherein
 in response to input of an image, the estimation model estimates, for at least one attribute value, a probability that the image has the attribute value. 
 
     
     
       9. A method, comprising:
 obtaining a plurality of images corresponding to respective values of at least one of dosage, temperature, and exposure time of a drug that affects occurrence of a reaction of the drug; 
 generating, for each of a plurality of thresholds relating to the at least one of the dosage, temperature, and exposure time, a learning data set in which at least a part of the plurality of images is each given an attribute value, the attribute value varying corresponding to whether or not the at least one of the dosage, temperature, and exposure time exceeds the threshold; 
 performing, for each generated learning data set, learning processing on an estimation model that estimates the attribute value of an image in response to input of the image; 
 outputting, for each estimation model, an evaluation value according to an estimation error of the estimation model; and 
 determining a value of the at least one of the dosage, temperature, and exposure time at which the reaction of the drug occurs based on the threshold, among the plurality of thresholds, corresponding to the estimation model having the smallest evaluation value. 
 
     
     
       10. A non-transitory recording medium having recorded thereon a set of computer-readable instructions that causes a computer to perform operations comprising:
 obtaining a plurality of images corresponding to respective values of at least one of dosage, temperature, and exposure time of a drug that affects occurrence of a reaction of the drug; 
 generating, for each of a plurality of thresholds relating to the at least one of the dosage, temperature, and exposure time, a learning data set in which at least a part of the plurality of images is each given an attribute value, the attribute value varying corresponding to whether or not the at least one of the dosage, temperature, and exposure time exceeds the threshold; 
 performing, for each generated learning data set, learning processing on an estimation model that estimates the attribute value of an image in response to input of the image; 
 outputting, for each estimation model, an evaluation value according to an estimation error of the estimation model; and 
 determining a value of the at least one of the dosage, temperature, and exposure time at which the reaction of the drug occurs based on the threshold, among the plurality of thresholds, corresponding to the estimation model having the smallest evaluation value.